Seyyed Mohammad Reza Hashemi - Faculty of Computer and Information Technology Engineering, Qazvin Branch, Islamic Azad University, Qazvin, Iran1
Ali Broumandnia - University, Tehran, Iran

With recent advances in imaging technology, digital images have been introduced as an important part of media distribution. Although usually presented once, images need to be altered under various circumstances for reuse. In addition, varying screens can be employed to show the images, which include high-resolution monitors to low-resolution mobile devices. As a result, in recent years, image resizing has gained much account, with methods being presented which are based on efficiency accuracy in resizing algorithm. This paper compares some new methods of resizing algorithm in standard MIT RETARGETING database. It has used a quality metric to evaluate the quality of image resizing, and has employed structural features of the image in SIFT detector as well as EMD metric to evaluate color quality of the resized image, eventually to show the efficiency of each in accordance to the results. Finally the paper gives a general conclusion of the new methods in this algorithm.

object detection, image salience, image resizing algorithm, Markof Algorithm